Design — Guilds (the namesake pillar)

High-level design for guilds — the game’s namesake and the fourth dimension of power, intertwined with the council. A guild has two halves: a plentiful license (bought, gates a craft) and a single scarce guildmaster seat (won by economic dominance, pulls craft-wide levers). Answers ../GAME_DESIGN.md §5 (power, the guilds pillar). Builds on positions.md (license & guildmaster are positions), production.md (crafts, building levels, the upgrade gate), politics.md (the council coupling + guild-recall), and economy.md (the profit metric). Design doc — the model and its rules, not an implementation plan.

Setting frame: the Crown charters the guilds; the guilds regulate the crafts and bankroll the politics; the council governs the town — and a dynasty muscles up through the guilds to reach the council. Keep it simple and fun: a sink, a gate, a bonus layer, and one more contested ladder — not a bureaucracy sim.

1. What a guild is

A guild is a town-level institution governing one broad craft-grouping. There is one guild per grouping (§2), shared by the whole town — not a per-dynasty thing.

Each guild has exactly two halves, mapping onto the locked two-tier scarcity (../GAME_DESIGN.md §5):

• The license — plentiful (§3). Any dynasty can buy a license to practise a craft. It is a positions.md position (seated in a living family member) and a gate: no license → you cannot build/upgrade that craft’s workshops or collect its bonuses. Just a matter of gold; nobody is hard-locked out of a craft.

• The guildmaster seat — scarce (§4). One seat per guild, the contested prize. It is not bought — it is won by economic dominance of the craft, held ex officio by the winning dynasty’s Elder, and it pulls the guild’s craft-wide levers (§5).

This is the fourth contested arena. Crucially it is contested in a different currency than the council: the council is won by social/political maneuvering (politics.md §3 — charisma, bribes, campaigns), the guild is won by out-producing rivals (§4). The two ladders interlock (§6) — that coupling is the whole point of the pillar.

2. The guild roster — four broad guilds

Rather than one guild per craft (which would make the guild count chase the ever-growing craft list), crafts are grouped into a small fixed set of broad guilds. A new craft added later simply slots into an existing guild — the roster stays fixed while production grows.

First-draft grouping, modelled on The Guild 2’s four character classes (revisable — a sensible starting cut, not a locked taxonomy):

• Building→guild assignment is deferred to the production.md craft split (which isn’t finalised), exactly as the craft taxonomy itself is. This doc fixes the guilds; production fixes which building lands in which.

• The Rogues’ Guild is real but crime-dependent. Its member “crafts” are mostly services, not crafting (pickpocketing, theft jobs), and they come online with the Crime pillar — now designed in crime.md. There, those jobs are the economic-crime category: tasks a Rogues’ den posts autonomously, whose net gold feeds this guild’s profit metric (§4) exactly like any craft’s.

3. The license — the plentiful tier

A license is the bought half, and it honours the locked promises (../GAME_DESIGN.md §5):

• It is a position (positions.md) — assigned to a living family member, and it expires when that holder dies → a costly rebuy + re-level. Death drama, reusing the vessel pattern wholesale.

• It gates production. No license for a craft → that dynasty cannot build or upgrade that craft’s workshops, nor receive its bonuses. This is the gate on the existing building-upgrade system, and a gold sink.

• It levels up. Contributing gold/materials raises a license’s level → more bonuses. Two-axis leveling with building level lives in production.md (building level × license rank).

• Bonuses are permanent-while-licensed craft buffs, plus purchasable timed “document” buffs — all just modifiers through the characters.md pipeline. (Document specifics parked, §9.)

• It does not count toward overload. A license is a guild position, and per the overload rule (positions.md §1) only dynasty positions (cabinet offices + building managers) pool overload. Licenses are meant to be plentiful — making them eat overload would contradict “just a matter of gold.”

4. The guildmaster seat — the scarce tier

The contested prize. One seat per guild, and it is won, not bought:

• Awarded by economic dominance. On a monthly tick, each guild’s seat goes to the dynasty with the highest summed net profit across all its shops in that guild’s category — every craft the guild spans, summed together. You earn the Craftsmen’s Guild by out-earning rivals across all your smithies and carpenters and tailors at once: a great craftsman, not just a smith.

  • Profit = net gold into the dynasty treasury over the month (revenue − inputs − wages), uniform across every building type — a sold sword, a sermon, a healing, a pickpocketing job all count the same. The shop manager already tracks this (controllers.md §7).

  • Hysteresis dampens the hand-over so the seat doesn’t thrash month to month — a challenger must clearly overtake the incumbent, not win by a single coin. (Exact margin: balancing, §9.)

  • Cadence and metric are balancing knobs — monthly is the starting point.

  • Held ex officio by the Elder. The seat is not seated in an arbitrary member; it rides the winning dynasty’s Elder (positions.md §3). This concentrates guild power on the seat that already writes the Agenda, runs dynasty-advancement, and is the rival-AI socket — so the political payoff of economic dominance lands exactly where strategy is decided.

• Succession is free. Because the seats ride the Elder office, they pass to the oldest member with it on the Elder’s death (positions.md §6) — no separate guild-vacancy handling. The live contest is a rival overtaking your profit and taking the seat at the next monthly tick.

• Does not count toward overload (§3, positions.md §1) — the Elder may hold several guildmaster seats (dominating several crafts) without penalty; the levers are dynasty-wide modifiers, an institutional title, not a body doing a job.

• Empty if uncontested — if no dynasty operates in a craft, its seat sits empty and its levers lie dormant.

5. The guildmaster’s levers

Holding a guild’s master seat lets the Elder pull that craft’s levers. A small starting set (more can be added once there’s gameplay to balance against):

Levers 1, 2, 5 are economic (squeeze rivals’ margins, protect your own); levers 3 and 4 are the bridge into politics (§6).

6. The two arenas — guild ↔ council coupling

This is the pillar’s reason to exist (../GAME_DESIGN.md §5):

  ECONOMIC FRONT                          POLITICAL FRONT
  out-produce rivals ──► guildmaster ──►  lever 4: sway council elections   ┐
  (monthly profit)        seat (Elder) ─► lever 3: recall an immune seat    ├─► the Crown
                                                                            │
        ▲                                 council legislates over guilds ───┘
        └──────────────  (charter / tax / override — the council's side) ───

• Up: economic dominance → a guildmaster seat → levers 3 & 4 → leverage over the council. The path to the Crown runs through the guilds.

• Down: the council legislates over guilds — charter, tax, override. The council’s specific powers over guilds are the council’s side of the coupling and are parked for a politics second pass (a returning office, or a Mayor power); this doc names the hook, politics designs the lever.

• Two contested arenas, different currencies, feeding each other — the flywheel of ../GAME_DESIGN.md §5, now concrete. Symmetric: rivals buy, level, and fight over licenses and guild seats too.

7. Already in code

Nothing. Guilds are 100% greenfield — zero references to guild/licen[cs]e in source/ or data/. A clean slate. What the pillar will reuse rather than build:

• The position/vessel pattern (positions.md) — license and guildmaster seat both ride it (member-seated, trait-tinted, death-vacated).

• The building-upgrade system (BuildingUpgradeSystem) — the §3 license gate hangs off the existing build/upgrade path.

• The profit ledger — the §4 metric reads the per-shop net-profit the manager already optimises (controllers.md §7); a city ledger (LedgerComponent / CitySystem) exists to extend.

• The modifier pipeline (characters.md) — every bonus, fee, and vote-modifier is just a modifier.

8. ⚠ Tensions & reconciliations

• Overload rule narrowed (reconciliation). The locked decision: overload is a dynasty-internal pressure only — it pools across cabinet offices + building managers (the seats you assign). Council seats and guild positions are exempt and do not interact with it or each other. This reverses the earlier text in positions.md §1/§5 and politics.md §5 (“council seats count toward overload”), which are being updated to match. Rationale: winning external contested seats should never punish you.

• Fully greenfield — licenses, the upgrade gate, the guildmaster award tick, every lever, and the council coupling are all unbuilt.

• The §4 award needs a monthly per-dynasty, per-guild net-profit total. Today’s ledger deducts/credits gold but does not bucket profit by craft category over a month; that aggregation is new.

• The upgrade gate inverts a current assumption — BuildingUpgradeSystem upgrades any owned building; under §3 it must first check the dynasty’s license for that craft.

• Rogues’ Guild waits on Crime, and Scholars’ service profits wait on the services design (production.md §1) — both produce real profit numbers, but the buildings that generate them aren’t built yet.

9. Open questions (next pass)

• Building→guild assignment (§2) — settled with the production.md craft split; and whether the four-guild grouping survives contact with the real craft list.

• License level count & bonus curve (§3) — how many ranks, and what each grants.

• “Documents” (§3) — the purchasable timed buffs: catalogue, costs, durations.

• Guild-recall coalition threshold (§5, politics.md §4) — how many of the four masters must agree to force a recall.

• The council’s side of the coupling (§6) — the council’s legislative powers over guilds (charter / tax / override); deferred to a politics second pass.

• Fee tiers, hysteresis margin, award cadence (§4, §5) — the balancing numbers.

• Special recipes (§5) — the parked lever; revisit when there’s gameplay to tune.